INDEX

Explanations

phrases indicating events, gatherings, or activities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÅĻÃŃ

-0.15

kÃ³w

-0.14

inton

-0.14

 crack

-0.14

igo

-0.14

Dul

-0.14

bara

-0.13

ise

-0.13

DU

-0.13

olic

-0.13

POSITIVE LOGITS

assis

0.16

 Stokes

0.15

.ua

0.14

uum

0.14

.va

0.13

/testify

0.13

 tetas

0.13

_strlen

0.13

skirts

0.13

_viewer

0.13

Activations Density 0.066%

phrases indicating events, gatherings, or activities

No Comments

No Known Activations